Cyberseminar TranscriptDate: May 10, 2017Series: Evidence-based Synthesis ProgramSession: Comparative Clinical and Economic Effectiveness of Anti-Vascular Endothelial Growth Factor AgentsPresenters: Glenn C. Cockerham, MD; Devan Kansagara, MD, MCR; Deborah Khachikian, PharmD; Allison Low, BA

This is an unedited transcript of this session. As such, it may contain omissions or errors due to sound quality or misinterpretation. For clarification or verification of any points in the transcript, please refer to the audio version posted at http://www.hsrd.research.va.gov/cyberseminars/catalog-archive.cfm

Molly: And I do want to kick things off now. So I would like to introduce our speakers for today. Presenting the results from the report, we have the lead author, Allison Low. She’s at the Evidence-based Synthesis Program Center, at the Portland VA Medical Center. Joining her as a co-author is Devan Kansagara. He’s a doctor, sorry, a director at the Portland Evidence-based Synthesis program, while also a physician of internal medicine at the VA Portland Health Care System and finally, an Associate Professor of Medicine, and Associate Professor of Medical Informatics and Clinical Epidemiology, and that’s at Oregon Health and Science University. Joining us also today as an operational partner discussant, we have Dr. Glen Cockerham. He’s the National Program Director for VHA Ophthalmology Service. And finally joining us, we have Deb Khachikian, and she’s the National PBM Clinical Pharmacy Program Manager at the Department of VA, Pharmacy Benefits Management Services. So at this time, I would like to kick it over to Devan to get things started.

Dr. Devan Kansagara: Okay, hi everyone. I’m Devan Kansagara. I’m just going to take one minute to introduce our program. So we had a lot of collaborators and co-authors, including our team. We had two ophthalmologists here on our team, and this topic was requested by Dr. Cockerham, who’s joined us today with the VHA Ophthalmology National Office, and we had a variety of technical experts. So just a word about the Evidence-based Synthesis Program, I would skip this, it’s a QUERI funded program. And we do systematic reviews on a broad variety of topics and use methods very similar to what the AHRQ EPC Program does. There’s four centers throughout the U.S., they’re shown here on the map. And this slide just gives you a quick overview. We try to work closely with our operational stakeholders. There’s a nomination process in case anybody on the call is interested in nominating a topic in the future. The link to the topic nomination form is at the bottom of this slide. And so, in this slide shows you the link to the full-length report on the ESP website, which contains links to all our reports, as well as links to slide decks if there had been a webinar and other papers or presentations associated with the topic. And so I am going to now turn it over. So Alli Low is our, one of our research assistants here, has been with us for five years and I just wanted to give her a shout out. She did the lion’s share of work, and all of the heavy lifting on this report. It was a tremendous

effort and she will be going off to medical school in two months. We will miss her and expect great things. So with that, I’ll turn it over and she’ll go through a couple of poll questions here.

Allison Low: Yeah, so thank you, Devan. So before we get started, we kind of just want to get an idea of who is on the line with us. So I’m going to hand it over to Molly to allow the poll. If you could just choose…

Molly: Go ahead.

Allison Low: Yeah, so if you could just choose, which of these best describes your primary role. That would be great.

Molly: Looks like we’ve got a nice response of audience, about just over half have replied. So we’ll give people just a few more seconds. So if you can just click your primary role, you can just click the circle right there on your screen next to your response. I understand that many of you wear many hats within the organization and we will have a more extensive list of job titles at the end in the feedback survey, so you can probably find your correct job title there if it’s not listed. Okay, we are at about 70% response rate, so I going to go ahead and close this out and share those results. As you can see, an overwhelming percent of our respondents, 89%, are VA clinician and 11% selected other. So thank you, once again. And now I’ll turn it back over to you.

Allison Low: Great, so now just one more question since the majority of you are clinicians. We are just wondering, which of these agents you use most often for ophthalmologic indication.

Molly: Actually, I’m sorry, Allison, can I get you to go through a few more slides and then we’ll come back to this poll question? I see I made an error in it, so I want to re-do it real quick.

Allison Low: Sure! No problem. Alright so now, I’m actually going to turn it over to Dr. Cockerham to just give us a little of background information on this topic before we share the results of our report.

Dr. Glenn Cockerham: Good morning, can you hear me?

Molly: We can, thank you.

Dr. Glenn Cockerham: Well I just want to give a little bit of background for, particularly for those who are not clinical ophthalmologists and don’t use these agents regularly. It’s an interesting story how these developed and how we currently are where we are with the agents. I’ll just start with the one that’s going to play a common role in the discussion today, particularly about cost and utilization. And this is the oldest drug, this is Bevacizumab, known as Avastin, which is continued to be used by oncology, and it got FDA approval for oncologic use in 2004. And because it was oncology and not intended for any place else, it was just a standard murine-derived monoclonal antibody. It happened to have two vascular endothelial growth factor binding sites. It was not felt, based on the way it was manufactured, in fact, it was murine in

size of it, that it would be a good choice for intraocular use, and so the company didn’t even ask for FDA approval at that time for it. However, it turns out that ophthalmologists realized the potential and started using it, since we had nothing else, we started using it off label, which is allowable, in 2005 in AMD, that stands for age-related macular degeneration, one of the leading causes of blindness, which is a slide later, I guess the way this is set up. And ever since that time, it’s been used worldwide, as you’ll can see, again, in the statistics we’re going to discuss today, it’s been used all over the world and in the VA as well. Next slide, please.

Molly: If you want, Glenn, I can launch that second poll now or we can continue on, your call.

Dr. Glenn Cockerham: Sure. Well, I might as well just do, I’ve got three slides on the agents and why don’t we do the poll.

Molly: And then we’ll go back to it. Okay, no problem. Go ahead.

Dr. Glenn Cockerham: Let’s just continue. So this is the same company. This drug is called Lucentis, and it’s Ranibizumab, again it’s a monoclonal antibody, as you can see by the generic name there. And this was designed specifically by the company who realized the value of anti-VEGF, particularly in the eye. And you say, why the eye, why do all these drugs matter in the eye? Well, the eye is a very compact organ and it’s got tremendously high blood flow. And the blood flow is necessary to maintain vision, because generating vision through the photo receptors and the supporting cells takes an intense amount of energy. And you need a high blood flow, both to bring the nutrients and to take away molecules in support of that. So there’s a very vascular choroid in the back of the eye underlying the retina. And so we have a high blood flow and we have a very compact space to work with, so a little bit of edema or bleeding in the eye can have a dramatic effect on vision. So the company realized that, they had ophthalmologists on their team, like [xxxx, 8:44], here in San Francisco, helped develop this drug. He had worked with [xxxx, 8:49] who developed the anti-VEGF originally in Boston. Anyway, so they developed this drug and got, went through the whole regulatory process to get FDA approval in 2006, originally for macular degeneration and then for other vasculopathy, vascular diseases in the eye. And this is considered to be a safer molecule, or at least it was considered to be a safer molecule, because it’s smaller, easier to get into the eye and it’s humanized. The murine component is minimized in this. And the company, obviously, did not want to compare the two drugs to each other, because one, as you will see is much cheaper, the [xxxx, 9:30] versus the one that was manufactured specifically for eye use, i.e., this one. But the NIH did do a head-to-head study called the CATT study of Lucentis versus Avastin which showed equivalency, which we’ll discuss I’m sure in the discussion today. And the third slide please.

And more recently, Regeneron has developed Eyllea (Aflibercept) and this is the most recent approval. It was approved in 2011 for age-related macular degeneration and then later for other things. And at this point, I think it would be a good time for the poll.

Molly: Thank you. Alright so for our attendees, as you can see on your screen, we do have the second poll question up. So for those of you who do use these anti-VEGF agents, which one do you use most often? Please choose one of the following. Oh goodness, Alli, can I get you to read these aloud, the answer options?

Dr. Glenn Cockerham: I can do that if you like.

Molly: Okay, if you have them up, yes, please do.

Dr. Glenn Cockerham: Yeah, so the first choice is Eylea, it’s the one that was most recently approved, Aflibercept; the second one is, in no particular order here, the second one happens to be Avastin, which is the oncologic drug that was used off label but still today, it’s never had FDA approval but it’s continued to be extensively used off label; the third drug is developed by Genentech/Roche and Novartis, which is Ranibizumab or Lucentis; the fourth choice is, I use all three and more or less equally; and the fifth choice is, I don’t use any. So it says, which do you prefer, the three versus I use them equally versus I don’t use them at all.

Molly: Thank you. So, 17% of our respondents said that they use Eylea and then the rest, 83%, said it’s non-applicable or they do not use these drugs. So thank you, to those respondents and we’re back on the slides.

Dr. Glenn Cockerham: Okay, so I’ll continue now. Let’s talk about the background then how can we describe the available agents. Again, we are talking about the vascular tissues in the eye that subserve vision and the conditions that can cause blindness due to growth of new blood vessels, which is called neovascularization. They cause blindness through bleeding, just of blood, and we’ll see a slide with just blockage of the visual cells that actually block them from seeing light and vision. Or leakage of the fluid into that compact space, which distorts the photo receptors and causes blurring of vision and distortion. You can also, if you’re leaking fluid, you can also leak lipids, which is called exudates, which also can block vision. And the three common conditions for approval that we have, the ones that have approval, are age-related macular degeneration, which we’ll discuss, diabetic retinopathy and retinal vein occlusion. The next slide, please.

Age-related macular degeneration is the most common cause of blindness in adults over 65 years of age. There are some factors that correlate with that including smoking and, of course, advanced age, hypertension. The one that really worries us is the one that leaks. The average macular degeneration that we routinely see is slide left. This is a [xxxx, 13:18] photograph. It’s a picture of the back of the eye, and you see, I don’t have a pointer, but you see the optic nerve and you see the blood vessels radiating out from the optic nerve on the left side of the left-hand slide. And within those blood vessels in the center is where the macula is, where it’s an avascular area that has the most density of photo receptors and sharpest vision. You need that for 20/40 or better vision. And unfortunately, because of the structure of the tissues, that’s where fluid tends to accumulate. So on slide left, we see, we see, the macula area, it’s got a lot of white flecks in it. Those white flecks are not normal, they’re called drusen. These are

lipofuscin deposits, which show that there’s a degeneration going on in the retina. And at some point, in 10% of the patients statistically, there’ll be bleeding in that. So on slide left, even though it’s not a high magnification, you see a spot of red blood and to my eye, I also see edema there, but you won’t necessarily see it. Slide right is a fluorescent angiogram where we inject fluorescein dye into the vessels to look for leakage, and we see that the leakage shows up as bright white. So we have leakage there, and this is not necessarily the same case, but it’s, I don’t think it’s the same, but anyway, you see that there’s a lot of blood blocking. The black is bleeding and the white is leakage, so this is what happens. And so, 10% of patients have that type of AMD and that is visually devastating. And it’s a large number, so about 200,000 new patients will get this yearly. And we didn’t have a very good treatment before this, but with anti-VEGF shots, we can get a lot of the leakage to go away and eventually the bleeding, hopefully, will be reabsorbed. Next slide.

Another common use is diabetic retinopathy. So we said that age-related macular degeneration is the most common cause in older people. This is the most common cause in younger people and most of you know the risk factor, or the correlations with diabetes mellitus. So this is the common cause of blindness in everyone between the ages of 20 up to 65. Macular edema hemorrhage again, in addition, you get proliferation of blood vessels because of ischemia. So the same condition that causes the leakage in the blood vessels is caused by VEGF, excessive vascular endothelial growth factor, which is produced in response to tissue ischemia. So this is, as far as the eye is concerned, this is a normal pathophysiologic response to generate VEGF to grow vessels, that’s what it does elsewhere in the body. It just happens to be in the eye and is visually devastating when it happens. So it’s the abnormal blood vessels that you can see there with the optic disc with all this pink trauma like growth on it. Those are aberrant or abnormal blood vessels that are growing in response to vascular endothelial growth factor, which has been produced. It’s produced by a whole variety of cells in the retina. Almost any cell can do it, predominant cell would be the vascular endothelium, but other cells, glial cells, and other retinal cells can do it as well. So there’s a lot of out pouring of VEGF in a condition like this that is producing new blood vessels. Next slide, please.

I don’t know if we want to go into this at this time. Do you want to continue in this or do you want to go… Shall I continue?

Allison Low: We can keep going.

Dr. Glenn Cockerham: Okay. So let’s just continue with how we actually treat it. So this is an office procedure, which is just done in any office. It’s done in the VA, in the room, we may have a special room where we do injections just for ease of flow and efficiency and so forth. It doesn’t have to be a clean operating room is the point. It can be done in outpatient settings and outpatient rooms. And it’s just done with topical anesthesia, believe it or not, all we use are eye drops to numb the outside of the eye, and we take a 30-gauge needle, which is what you see here, and slide right shows an actual procedure. We do wear gloves, but we don’t go full surgical gown or anything. We do prep the area and we do put drapes around it, sterile speculum, and again gloves, but again it can be done in the office. And we put the 30-gauge

needle into the vitreous cavity as you see on slide left. And the needle is just injecting the anti-VEGF agent into the vitreous cavity. And by doing that, you don’t have to put it underneath the retina or anything, it just goes into the vitreous gel and through diffusion it gets into the areas beneath it. Next slide. Okay.

Allison Low: Great, thank you so much, Dr. Cockerham, for that background. Here we just wanted to mention that we developed the scope and key questions with Dr. Cockerham, as well as our technical expert panel, which Devan described earlier. And we conducted the systematic review according to standard methodology.

The key questions that guided our report where basically, what are the comparative effectiveness, the comparative harms, and the comparative cost-effectiveness of these agents. We conducted a search in coordination with the research librarian, as well as searching some trial registries. And we also requested some unpublished data from the drug manufacturers. And we did receive responses, but none of the data was anything new that was not already in the published studies.

So the populations we looked at were the same ones that Dr. Cockerham just described. And in terms of study designs, we only included head-to-head randomized or non-randomized controlled trials that compared any of the three agents. But for this third key question on cost, we also expanded the study design to include things like cohort studies, or validated modeling studies, although we did limit them to studies in the U.S. because of the variability in drug cost across countries.

The main effectiveness outcomes we were interested in related to visual acuity. And these were usually reported as best corrected visual acuity or, BCVA, according to the chart developed by the Early Treatment of Diabetic Retinopathy Study. Specifically, we looked at mean change in visual acuity from baseline to follow-up. And based on consultation with our technical experts, as well as research on what visual acuity differences are actually able to be perceived by patients, we set the threshold for what would constitute a clinically meaningful difference between groups at five letters. And another major effectiveness outcome was the proportion of patients with an improvement in visual acuity of fifteen or more letters. We also looked at functional status or quality of life outcomes, although they were not commonly reported. And we also looked at intermediate outcomes such as those seen by anatomical changes seen through OCT imaging, but we’re not really going to discuss those today because they didn’t have a meaningful contribution to the results.

Here we just wanted to show you an example of the chart we are going to be talking about. You might be more familiar with the Snellen chart, which uses a 20/20, 20/40 system, but with the ETDRS chart, each line only has five letters and the patient just reads down the lines until they reach a row where three or more letters cannot be read, then their visual acuity score is the correction number of letters they were able to read.

For the key questions on harms, we looked at both ocular and systemic adverse events. For those who might not be familiar with this topic, the potential for increased risk of systemic harms is due to the fact that some of the drug from the injection enters the blood stream, which could reduce the body’s systemic levels of VEGF. And one of the most commonly reported of this type of event was the composite outcome of arterial thrombotic events, which comes from the anti-platelet trialist collaboration. And this is defined as vascular death, non-fatal myocardial infarction, or non-fatal stroke. And for looking at cost, we were obviously interested in the actual cost of treatment, but since that is not commonly reported in these studies, we also looked at proxies for costs, such as the number of injections and also those who needed other types of rescue treatments.

We rated the quality of each included study and we also pooled the results using meta-analyses whenever it was possible and appropriate to do so. We also rated the overall strength of the body of evidence for the outcomes by taking into account several factors, such as the quality of the evidence, limitations of the studies, consistency and precision, as well as directness, which relates to applicability.

This literature flow diagram just summarizes the results of the search and study selection process, which resulted in sixteen eligible studies. And these studies were fairly diverse. They ranged from studies with treatment arms as small as 8 to over 300 patients. And they also used a variety of treatment schedules, which included both monthly injections, as well as PRN or as needed dosing with varying retreatment criteria. So 11 of the 16 included studies were conducted in patients with neo-vascular age-related macular degeneration. And again, many of the studies used PRN dosing. And the most common comparison was between Bevacizumab and Ranibizumab, which was reported in 9 trials. And these trials reported no significant differences between the agents and visual acuity outcomes, which we’ll take a closer look at on the next slide. Nor was there a difference in the quality of life in the one trial that reported this outcome.

So first, I’d just like to orient those who might not be as familiar with these forest plot diagrams. So this vertical solid line at zero represents no difference between arms. And each line or row represents a separate study with the box representing the point estimate and the line denoting the 95% confidence interval. And the size of the box kind of corresponds to the size of the study. If the confidence interval crosses this line of no difference, it means that there was no statistically significant difference between the arms. And this blue diamond here at the bottom represents the summary estimate with this, kind of, vertical axis with the red dotted line signifying the point estimate, and then the two lateral sides of the diamond corresponding to the confidence interval. So for this plot specifically, which represents the mean difference in letters changed between the Bevacizumab and Ranibizumab groups over 12 months. You can see that the summary estimate clearly crosses this line of no difference, but also each individual study also showed no significant difference between the drugs. And just as a reminder, as we mentioned earlier, we classified the clinically meaningful difference as a difference of at least 5 letters. So obviously, this is very much below that threshold. We also just wanted to mention that the forest plot for the four trials that reported longer term results had similar findings.

This is the plot of the other major visual acuity outcome, which is the proportion of patients in each group gaining 15 or more letters. And again, the pooled estimate shows no significant difference between the groups.

For the comparison of Aflibercept versus Ranibizumab in this population, there were two identical trials that were conducted in different countries. Each trial had four arms, three of which were different doses and injection schedules of Aflibercept, which were compared to monthly Ranibizumab. And the results of these trials were conflicting with one trial finding a small benefit with Bevacizumab and the other with Ranibizumab, but the difference for both trials was less than this five letter difference, so it was not considered clinically meaningful. And actually, when we tried to pool these studies in a meta-analysis, the statistical heterogeneity was very high, so we were unable to pool them for a summary estimate. So therefore, we give this outcome an insufficient strength of evidence rating, meaning that more evidence is needed in order to have any confidence in the estimate. And one finding mentioned by these trials is that Aflibercept injections that were administered every other month, after three initial loading doses, had similar efficacy to monthly Ranibizumab. So it’s possible that even though Aflibercept was not found to be superior in this trial, you might be able to get away with fewer injections. Although, it should be noted that since there was no arm that was given bimonthly Ranibizumab, it’s possible that that would have resulted in similar visual changes as well. And there were no trials comparing Aflibercept to Bevacizumab in patients with AMD.

So in patients with diabetic macular edema, the bulk of the evidence came from the large DRCR trial, which compared all three of the drugs. In comparing Bevacizumab to Ranibizumab, there was moderate strength evidence of no difference in visual acuity changes between the drugs, and none of the trials reported quality of outcome data, or quality of life data, sorry.

And in this plot showing the main difference in letters changed over twelve months, you can see that the DRCR trial provided the most precise evidence and none of the trials, nor the summary estimate were significant.

There was also no difference between the groups in the proportion of patients with the 15 letter or more improvement.

Comparing Aflibercept and Bevacizumab in this DRCR trial, there was a small benefit with Aflibercept, but the difference was less than five letters, and therefore probably not clinically significant. However, in a subgroup analysis in patients with lower baseline visual acuity, which was defined as less than 69 letters, the benefit with Aflibercept was found to be clinically significant at 12 months, since it was greater than five letters, but the difference was no longer significant by 24 months. And the results of patients gaining 15 or more letters was similar with some short-term benefits, but by 24 months Bevacizumab had caught up. And also in patients with better baseline visual acuity, there was no difference at either time point.

The comparison between Aflibercept and Ranibizumab followed a similar trend, with a small benefit with Aflibercept, but even in the subgroup analysis of patients with lower visual acuity, the difference didn’t reach clinical significance.

So for the clinical population of retinal vein occlusion, there were only two small short term trials that compared Bevacizumab to Ranibizumab, so we have no data on Aflibercept in this population. Both of these trials showed no difference between Bevacizumab and Ranibizumab, but we graded the strength of evidence as insufficient because the effect estimate was very imprecise, as you can see in this forest plot. There was also no difference in the percentage of patients gaining 15 or more letters, and neither of these studies reported quality of life outcomes.

Most of these included trials reported some harms data, but it varied from very limited to very detailed. And I should point out that none of these trials were specifically designed to assess harms, so they were likely not powered to detect a difference in adverse events between the arms, primarily because the rate of events was so low.

In patients with AMD, the rates of people withdrawing due to adverse events, as well as serious ocular adverse events were quite low, including endophthalmitis. Another event such as retinal detachment and hemorrhage, and there were no significant differences found between groups.

Systemic adverse events were more common in these trials. And while there was a fairly large range between studies depending on how they defined certain events, there were generally no differences found between the drugs. However, one exception is this CATT trial, which had the highest rates of patients with at least one systemic event, and found a greater risk associated with Bevacizumab. However, a large portion of these events were things that were not previously found to be associated with systemic anti-VEGF treatment for cancer, where much larger doses of anti-VEGF drugs are introduced into the blood stream. So it’s unclear, kind of, what to make of this finding. The other five trials found no difference between the drugs for overall systemic harms, but one trial actually found a higher rate of arterial thrombotic events in the Ranibizumab arm, but this difference disappeared by 24 months.

Ocular adverse events also occurred very rarely in patients with diabetic macular edema, with no significant differences reported between the groups. Endophthalmitis only occurred in one patient of 660 in this DRCR trial. And the rates were higher in the other two small trials from Turkey and Brazil. But in the trial that reported these occurrences by arm, both instances actually occurred in the Ranibizumab arm. As far as systemic adverse events, the DRCR trial reported significantly more arterial thrombotic events in patients treated with Ranibizumab compared to Aflibercept. Otherwise, although the rates of events were quite high, as might be expected by a fairly sick diabetic population, there were no differences found between the drugs.

Since there were only two small short term trials in patients with retinal vein occlusion, there’s not a whole lot we can say about harms in this population. However, the comparative rates of adverse events are likely to be similar to those found in the other population.

As far as costs, two large trials provide moderate strength evidence that treatment with both Ranibizumab and Aflibercept are substantially more costly than compounded Bevacizumab, and provide no incremental cost-effectiveness benefits. However, there was no data on the cost of non-compounded Bevacizumab, which I know we’ll talk about in a little bit as well. And this is where the large vial is used for a single patient with the rest discarded, which is the method currently used by the VA due to safety concerns.

The first trial was in patients age-related macular degeneration and reported on the cost of the drug itself per patient based on the average number of injections given over the 24 months of treatment, and found that even though the Bevacizumab group treated as needed was given more injections than the as needed Ranibizumab group, Ranibizumab costs 30 times more over the course of treatment because the per dose cost is drastically more expensive than compounded bevacizumab. And the DRCR trial used a broader definition of treatment costs that included not only the cost of injections, but also things like adverse events, and rescue treatments with laser photo coagulation, and still found that Bevacizumab was significantly less expensive. This trial also calculated quality adjusted life years per group based on average, based on adverse events and visual acuity achieved, and found that even the subgroup with the lower baseline visual acuity, which was again the group that saw a significant benefit with Aflibercept, the ten-year incremental cost-effectiveness was nearly $290,000 per quality adjusted of life year gained, which is well above the $50-150,000 threshold usually sited by the cost-effectiveness literature as acceptable.

There were several limitations over systematic review including the methodological limitations of the included studies, which included things like the lack of an intention to treat analysis, core reporting on harms, and insufficient methods of allocation concealment. And also, there were very few trials on patients and retinal vein, with retinal vein occlusion and only a couple of studies compared Aflibercept to the other drugs. And also, some of the trials didn’t enroll patients with certain cardiovascular risk factors, which may limit their applicability to many of the patients receiving these treatments. Another limitation is that only one of the trials decided to stratify the results by baseline visual acuity. So we don’t know if similar results would be found in other conditions or trials. And as previously mentioned, the fact that all of the cost data was based on compounded Bevacizumab, which the FDA actually currently has a draft guidance document out that details the rules for how long drugs like compounded Bevacizumab can be stored, which I believe in the current draft is only a few hours. But until the final version of this document is published, the feasibility of using compounded Bevacizamab In the future is unclear. There were a couple of other limitations that we talked about in the report. I’ll let you maybe look at those there because of time.

To briefly summarize our findings on effectiveness, we found low to moderate strength evidence of no consistent clinically meaningful differences between these anti-VEGF agents for

the general population, although the evidence was insufficient for retinal vein occlusion. However, one trial suggested that Aflibercept may be superior in patients with lower baseline visual acuity, at least over the short term, although the longer-term effects are unclear, and more trials are needed to characterize its potential benefits over the other agents, as well as the potential for a less frequent dosing schedule.

And in terms of harms, the rates of serious ocular adverse events were low and similar across studies. And no clear differences were found between the drugs and the rates of systemic harms. But treatment with compounded Bevacizumab was considerably cheaper than the other two agents, but we have no information on non-compounded costs. But in choosing amongst these drugs, clinicians may need to consider other factors such as patient preference, convenience, and distant to facility.

So I’m now going to turn it over to Dr. Cockerham to talk a little bit more on how this relates to the VA.

Dr. Glenn Cockerham: Sorry, my phone timed out, I’m back. Is Dr. Khachikian on the line from the pharmacy?

Dr. Deborah Khachikian: Yes, I am.

Dr. Glenn Cockerham: Perfect, I may need you to back up on this if we get into a discussion. The first slide is VA costs of anti-VEGF drugs and as we’ve continually referred to, the Bevacizumab, the original drug that was developed by Genentech for oncology, and it was intended to be injected intravenously, is relatively cheap, it’s come down quite a bit in the last decade. Again, this was I believe, got FDA approval in ‘04 for oncology. Anyway, it’s a large vial because it’s used for oncology and the VA cost is around $500 a vial. That being said, you can get at least ten doses if you break it up. Also being said, that the VA does not allow ophthalmologists to currently do that because of infection control. There’s issues with fungal endophthalmitis due to poor practices of compounding pharmacies that allow the vial to get contaminated, and if break, if you take a contaminated vial and you break it up and injected into ten or more eyes, then you’ve got ten infections. So anyway, that’s the cost for the large vial. Ranibizumab is, come down quite a bit for 0.3 mg. It used to be about $2000 a dose for 0.5 mg, now it’s $1400. And for the smaller dose for macular edema, it’s as you see it there. The newer drug at the bottom, Aflibercept 2 mg, generally you can get, you might be able to get two doses out of that, but generally it’s about one dose out of the vial. It’s $1400, it’s a little more expensive than using the smaller dose of Ranibizumab however, as we referred, or at least we did mention, it can be given bimonthly. Plus, the trial was monthly Bevacizumab versus bimonthly Aflibercept, and they were equivocal. So that is the advantage, so you could cut the cost in half and that’s why the acceptance is increasing with Eylea with a lot of retinal surgeons and ophthalmologists. Deb, would you like to comment on anything I’ve said so far?

Dr. Deborah Khachikian: No, just to point out that these are internal VA costs, so out in the rest of the world, their prices may be very different.

Dr. Glenn Cockerham: Okay, but these are fairly stable at this point, aren’t they? You don’t expect them to change much in the next year.

Dr. Deborah Khachikian: You know, they could change. The prices you see here have changed from the prices that we had last year. For example, there was a time, I think it was about a year or two ago, where the 4ml Bevacizumab vial was about $400, so it has gone up about $100 over time. So, yes, that’s certainly a possibility. Other times, companies want to compete and they come to the PBM and, you know, may offer special pricing, which can further reduce the prices. So it is somewhat, you know, it can move around, the price. But this is what we have for right now.

Dr. Glenn Cockerham: Okay. Next slide, thank you, next slide please. So how much do we spend? Again, courtesy of Pharmacy Benefits Management and Debbie Khachikian, this is the purchasing data from April, 2016 to March 2017, and you see as I mentioned, there’s increasing usage, well you don’t know, you don’t see the curve, anyway there’s a lot of usage of the Eylea, the one that’s used twice, once every other month. And Bevacizumab is used quite frequently, but it costs a lot less, so overall cost is not as significant as the Eylea. And Lucentis is kind of taking a back seat. Lucentis is Ranibizumab at two different doses, it’s taking a back seat to the other two drugs. However, that’s a bit confusing, so maybe Bevacizumab is not used quite as much as it looks like here by ophthalmology because it continues to be used in oncology [xxxx, 45:22] so I don’t think we have a way to break that out do we Deb, we just have the overall cost.

Dr. Deborah Khachikian: Yeah, that’s correct. That cost represents both potential uses. You know, what you don’t see on this slide though are the number of units purchased. So actually, for Aflibercept, we’re purchasing, that price represents approximately 53,000 units that were bought, whereas Bevacizumab, that price represents about 36,000, over 36,000 units purchased. So we are still using a lot of Bevacizumab, but just that the overall purchasing data appears to look as though, the least, not commonly used, but it’s just that the acquisition cost of that product is much less. And our Ranibizumab use, the number of units purchased were about 8,000 units, so that is the least used agent of the three in the VA.

Dr. Glenn Cockerham: Thank you. The next slide, please. So I refer now to the data on this that’s in the literature. Looking at Medicare costs and VA costs up to 2011. And this from Dr. Suzanne Pershing, who’s at Palo Alto. Suzanne, by any chance, are you on the line? She could explain her slide better than I can. But this was published in Health Affairs recently in 2015, looking specifically at age-related macular degeneration. Again, it’s Bevacizumab, Lucentis, and of course the dosing prices are perhaps different than the VA, but it was estimated anyway. What they found was that at that time through 2011, there was an upward sloped curve for Bevacizumab being used more frequently over Lucentis and we still see that. Now we’re 5, 6 years, 5 years later with the data we just looked at, let’s say. And we still see that bevacizumab is used more frequently that Lucentis, which is probably tailing off even more, at least in the VA. But since then, Aflibercept is becoming more frequently used.

Now one thing we don’t really need to get into since we’re VA, but there is, as Dr. Khachikian had mentioned, there are reasons why people do what they do besides the efficacy of the drugs because it doesn’t seem that the efficacy of the drug seems to matter. So what drives behavior and what drives utilization and what drives physician choices, there are incentives. In CMS, not so much, well in CMS, the patient has an incentive if the doctor uses something that’s less expensive because they have less of a co-pay. On the other hand, a physician has an incentive, at least in CMS, to get higher reimbursement for a more expensive drug. CMS will pay the physician if they have a practice or an office, 6% of the drug cost as an overhead. So I don’t know, it’s impossible to say, and I would like to think that financial incentives don’t play much role, but I just throw them out there for your consideration. Next slide, please.

So this is an article from 2014, again in Health Affairs. The title sums it up pretty well, “Switching To Less Expensive Blindness Drug Could Save Medicare Part B $18 Billion Dollars Over a Ten-year Period”. And estimated Medicare savings is said would be $18 billion and the beneficiary co-payment could be up to $5 billion. At that time, using the current prices at that time for Bevacizumab versus Lucentis, and Eylea was just coming on the scene probably, I think when this article was…so again it’s changed, but at that time, if you go head to head with Lucentis price at that time versus Bevacizumab, that’s the estimated cost savings. There’s still tremendous cost saving for Bevacizumab over the other two drugs. Next slide, please.

And that’s the last slide, so…

Dr. Devan Kansagara: Thanks, everyone. So, I guess we can do questions. Let’s see…

Molly: Okay, there are no pending questions at this time. For our attendees, to submit a question or a comment, please go to the Go To Webinar control panel on the right-hand side of your screen and click the arrow next to the word questions, that will expand the dialog box and you can then submit your question or comment there.

We do have a couple of people writing in saying, thank you, are these slides available? Yes, they are available. You have the link in your reminder email or you can write in to the question box and I can send you the direct link there. Another person writes that they’re thankful that you did this presentation and they look forward to sharing it with their ophthalmology team. But no pending questions per se. While we wait for any to come…

Dr. Devan Kansagara: I have a question.

Molly: Oh, go ahead.

Dr. Devan Kansagara: So I have a question so, and maybe Glenn or Debbie has insight on this. But it’s interesting given that the, you know, kind of data that Alli went through, why is Ranibizumab use falling out of favor? I mean there doesn’t appear to be data, kind of, suggesting it necessarily should be falling out of favor.

Dr. Deborah Khachikian: The way I interpret that is, this is Deb, the way I interpret that is that, you know, Bevacizumab and Ranibizumab are very similar products. And the cost of Ranibizumab, you know, is more, it’s twice that of Bevacizumab, at least for AMD and RVO. So it makes sense to me that one would use be Bevacizumab and that Ranibizumab utilization, you know, would decrease. In your own evidence that was presented here, the clinical trials, support that in that the outcomes from both an efficacy and safety standpoint were very similar between the two products.

Dr. Devan Kansagara: That’s true, yeah, I guess I was thinking more in terms of the Aflibercept versus Ranibizumab use. But, yeah, I guess in terms of, maybe it’s going down in favor of bevacizumab rather than of Aflibercept.

Dr. Glenn Cockerham: Well, yeah, and it, from the perspective of the patient, coming in every other month is enormously more attractive than coming in every month for an injection. I would minimize the number of eye injections that I had to get if I could. And it’s easier for the physicians and their office staff. And until new data comes out that that’s not a good practice then, I guess that trend is going to continue.

Molly: Thank you. Does anyone want to make any wrap up or concluding comments before we tie this up? And in no particular order, Deb, did you want to wrap up with anything?

Dr. Deborah Khachikian: No, I don’t have anything to add, thank you.

Molly: And Dr. Cockerham, was that it for you or did you want to add anything else?

Dr. Glenn Cockerham: I’ll just add one comment. There’s one other large organization that we didn’t mention, that’s the Food and Drug Administration. We did talk about FDA approval versus non-approval, but currently we are kind of hanging in the balance to see what the FDA is going to do. They still have not resolved the documents about the compounding issue, which we don’t do anymore in the VA at this time. It could change, who knows, if it might change in the future, but at this point we have a beyond use date once you repackage the Bevacizumab and you break it into smaller units so that you can inject it into multiply patients, that’s called repackaging, and it’s currently beyond use date of four hours. So once you open the vial, you’ve got four hours to do whatever you are going to do, and that doesn’t work very well in our current system in the VA at this particular…we wouldn’t get the full benefit of the drug that’s in the vial if we have to throw it away from the time it’s opened in the pharmacy to its last used in the clinic, so that’s another issue before us. And the FDA still hasn’t resolved what they are going to do. There is some talk about making it longer, but then they had some rules about cultures and so forth. You have to prove that your compounding pharmacy is compliant, so…that’s the only comment I had.

Molly: Thank you. And Devan and Alli, did either of you want to wrap up with anything?

Allison Low: Yeah, just kind of to dove tail on what Dr. Cockerham is saying. It does seem like there’s quite a bit of controversy between, kind of, the policy makers and the clinicians on the ground on that issue. I think several professional organizations when the original draft guidance was submitted and open for public comment, a lot of those organizations, kind of, tried to show it from their perspective and, kind of advocate for a longer beyond use date for these drugs. In order make it more feasible to use in clinical practice.

Other than that, we just want to thank everybody for participating. If you think of any additional questions, feel free to email any of us here on this last slide. And the full-length report is available online as well. So, thank you again.

[END OF AUDIO]